CHAPTER 4: Determining the population genetic substructure of Swakara sheep using
4.5 Discussion
Studies of genetic relationships within populations reveal useful information within and between populations that can be used for breed improvement programs (Kantanen et al. 2000). Problems of sub- vital performance are seen in some of the white Swakara sheep and it is hypothesized to be as the result of intensive selection for white pelt characteristics in the original Swakara population. The Swakara sheep of Namibia were introduced in 1907 from Germany and developed through years of selection and crossbreeding with breeds such as the Blackhead Persian and Namaqua. The Swakara sheep has continued to show superiority in terms of producing pelts with unique and distinctive characteristics that characterize the breed (Schoeman 1998). Years of intensive breeding can lead to inbreeding depression, a phenomenon where recessive genes with undesired phenotypes are expressed in the population and it is possible the sub-vitality observed in Swakara is a culmination of these factors.
The study population consisted of Swakara sheep collected from Namibia and South Africa and also six individuals from Germany. Assessing the genetic diversity between Swakara and Karakul would highlight how much the Swakara has been developed such that it is different genetically from the Karakul of Germany. The preliminary results indicated differentiation of the Halle population from the other colour subpopulations, which were therefore excluded for further analysis. The Swakara sheep belonged to different pelt colour subpopulations of white, grey and black. The white sheep were either of vital white or sub-vital white performance and were treated as two subpopulations whilst the grey and back were of vital performance. Using the OvineSNP50 beadchip with 54 241 evenly spaced SNP markers, genetic diversity within, the presence of population structure and the genetic differences amongst Swakara subpopulations was investigated.
The inbreeding coefficient (FIS), population differentiation and FST statistics were used to investigate the relationships amongst the colour subpopulations. The inbreeding coefficient measures the probability that an individual has two alleles at a locus that are identical by descent by comparing the frequency of heterozygotes in a population relative to the frequency of expected heterozygotes under random mating
(HE). F statistics in general measures the degree by which heterozygosity (genetic diversity) is reduced in a population under HWE causing populations to diverge. Genetic diversity and population structure are important; characterizing these two aspects in Swakara sheep will help with an understanding of the genetics of disorders such as sub-vital performance and this could lead to possible breed improvements at the genetic level.
The most gene diversity was in the grey population (HE = 0.3417) which had the lowest inbreeding coefficient (FIS) of 0.0092. The population with the lowest genetic diversity was the black population with HE of 0.3389 and FIS of 0.0496 followed by the vital-white population with HE of 0.3301. In the Karakul/Swakara pelt industry, coat colour is an important trait that has been selected for over years.
The black is the most abundant and readily available, while the white is preferred because the pelts can be altered to different colours and made into different garments. Selective breeding in most instances compromises the diversity of a population (Wiggans et al. 2010), which could have been the case with the white and black populations. Loss of genetic diversity can affect a population’s evolution and reduces individual fitness. Higher FST values were observed between the grey and white-sub-vital that had a genetic differentiation FST value of 0.6846 on chromosome 13. These results indicate that the SNP on chromosome 13 is highly differentially selected between the sub-vital white and grey. Overall, the genetic differentiations revealed observed on chromosomes 5, 6, 7, 10 and 13. Chromosome 10 had two SNPs with high FST values between the white-vital and sub-vital Swakara sheep (FST =0.4908) and between the grey and black populations (FST = 0.5209). These SNPs were checked for genetic functions or related genes and none were found. The other SNPs which showed high differentiation were also investigated for genomic functions and only the SNP on chromosome 6 with an FST= 0.5727 between the black and white-sub-vital sheep was associated to the GABRB1 (gamma-aminobutyric acid A receptor, beta 1) gene (Table 4.5), which is responsible for protein encoding in the central nervous system. A study on signals of selection was conducted on sheep to a SNP with a high FST in the global sheep study population was located close (29.5 Mb) to gene responsible for horns (RXPI2). Over the years breeders have been selecting against horns and this supported by the history of sheep breeding as
a strong signal of selection (Kijas et al. 2012). Only FST values of over 0.5 are considered for signals of selection in this population study and the SNPs that qualified showed no gene related any function.
Genetic structure was also determined using principal component analysis. The high eigenvalues of the first (36.89%) and the second (28.8%) principal component indicate the presence of non-random population structure (Figure 4.4). The principal components showed a clustering of the six individuals collected from Halle, Germany (Figure 4.4 D) separate from the rest of Southern Africa populations thus becoming an outgroup and over the years the Karakul and the Swakara have come to differ genetically. The individuals collected from Gellap-ost, Namibia (Figure 4.4 A) clustered in two groups with another group congregating individuals collected from private farms in Namibia (Figure 4.4 C).
This extent of genetic relatedness may be due to the fact that the Swakara industry in Namibia is consolidated (Louwrens et al. 2004) and therefore the same genetic pool is used to create different flocks. The population collected from the Northern Cape also formed a distinct cluster (Figure 4.4 B) that contained individuals from Gellap-ost, showing some genetic similarities among some individuals of the two geographical populations. The Swakara populations would likely have had less genetic gain and as such flocks are likely to have some genetic similarity (Groenewald et al. 2010). The breeding history of a flock also contributes to its population structure and the divergence between individuals from different stations could be due to isolated breeding and the difference in breeding management between stations. The second PCA separated populations according to pelt colour. The white-vital and some of the white sub-vital sheep clustered together forming an isolated cluster around 0.15 and -0.1 coordinates (Figure 4.5 A). The other white sub-vital individuals also formed a second cluster together with some of the black individuals (Figure 4.5 C). The grey individuals were dispersed over a wider admixed cluster with traces of black and sub-vital individuals in that cluster (Figure 4.5 B).
FST and PCA-based results revealed that the vital-white subpopulation was more genetically similar to the grey population and more distinct from the black subpopulation. By comparison the white subpopulation shared a selection signal with the sub-vital white population at chromosome 10 that had
an FST value of 0.4908 but the SNP (OAR10_87219588) had no associated gene function to it. The white and grey subpopulations shared an FST value of 0.3772 for a SNP (s53466.1) on chromosome 7 but had no function or gene associated to it. The highest fixation index (FST = 0.6164) between the subpopulations that shared the least genetic similarity was on SNP OAR13_16487810.1 on chromosome 13 that had no gene function associated to it.
This supports the within subpopulation genetic diversity analysis that revealed low genetic diversity in the white and black subpopulations. Highly inbred or least diverse populations will be more diverged from other populations because of the reduced allelic variation within them. The South African Karakul industry was developed from flocks taken from Namibia (Campbell 2007) and therefore would be expected to be more genetically similar than the Halle group. The analysis based on location showed a more hierarchical structure and admixed clusters were observed in which the Gellap-ost, private farms and Northern Cape individuals were grouped together. In a study of global sheep populations, PCA analysis grouped individuals of same breed and geographic location such as Africa, East Asia and North American breeds together (Kijas et al. 2009). In our study population we found that the Halle group diverged significantly from the Namibia and South Africa populations. A principal component analysis for the African populations that excludes the Halle group would be important in order to look at how much the South Africa and Namibia populations differ at a geographic location level.
Bayesian-based analysis showed some overlapping genomic regions between subpopulations particularly the vital-white and sub-vital white populations as well as the black and white sub-vital sheep. ADMIXTURE is a model-based clustering method that partitions the genome of each individual into predefined number of components. Population sub-structuring is examined using a clustering algorithm. The analysis for the population structure in Swakara sheep using ADMIXTURE was similar to that of PCA that revealed how closely related the subpopulations were to each other. The black and the sub-vital subpopulations shared some genomic regions that were different from the rest of the subpopulations. The Swakara population is more genetically similar amongst the subpopulations. The
extensive breeding on each subpopulation may have resulted in the admixture seen in Swakara population. The grey subpopulation also experiences the lethal factor that affects growth and development and as such it is usually cross-bred to eliminate the factor (Schoeman 1998; Rothauge 2009) which might explain the genetic diversity in this subpopulation due to the breeding practices i.e.
avoiding grey x grey mating.